Visa has been granted a patent for a system that analyzes multivariate time series data to predict contemporary transaction metrics. The system uses machine learning models with advanced processing layers like fast Fourier transform and dual-attention components for accurate predictions. GlobalData’s report on Visa gives a 360-degree view of the company including its patenting strategy. Buy the report here.

According to GlobalData’s company profile on Visa, Contactless Card Payments was a key innovation area identified from patents. Visa's grant share as of February 2024 was 43%. Grant share is based on the ratio of number of grants to total number of patents.

System for analyzing multivariate time series using machine learning model

Source: United States Patent and Trademark Office (USPTO). Credit: Visa Inc

A recently granted patent (Publication Number: US11922290B2) describes a system that involves a processor receiving historical data points and training a machine learning model to predict future transaction metrics. The system utilizes a fast Fourier transform (FFT) layer, a feature extraction component, and a dual-attention component to process the data. The processor combines real and imaginary parts of different features from historical and contemporary time periods to generate predictions for a target transaction metric during a specified period. The model is trained to provide accurate outputs based on the input data, allowing for effective forecasting of transaction metrics.

Furthermore, the patent details a method and a computer program product that follow a similar process of receiving historical data, training a machine learning model, and utilizing various components to generate predictions for future transaction metrics. The method involves determining time series data associated with historical and contemporary transaction metrics, processing the data through a series of layers, and generating attention matrices based on real and imaginary inputs. By combining different features and utilizing convolutional layers, the system aims to enhance the accuracy and efficiency of predicting target transaction metrics. Overall, the patent outlines a comprehensive system, method, and computer program product for leveraging machine learning techniques to forecast transaction metrics based on historical data points.

To know more about GlobalData’s detailed insights on Visa, buy the report here.

Premium Insights


The gold standard of business intelligence.

Blending expert knowledge with cutting-edge technology, GlobalData’s unrivalled proprietary data will enable you to decode what’s happening in your market. You can make better informed decisions and gain a future-proof advantage over your competitors.


GlobalData, the leading provider of industry intelligence, provided the underlying data, research, and analysis used to produce this article.

GlobalData Patent Analytics tracks bibliographic data, legal events data, point in time patent ownerships, and backward and forward citations from global patenting offices. Textual analysis and official patent classifications are used to group patents into key thematic areas and link them to specific companies across the world’s largest industries.